True peak

[nikkdust]

Another one of those threads...

Lets talk true peaks, for example why does spotify need as much as -1db true peak? Or even -2db, when mastered louder than -14Lufs. Is that really a hard rule to follow? Or do you give a d@mn about it?

I don't understand these things, I only create music

 

[SwitchBack]

Without getting too techy there's always chance of some overshoot with peak sample values, i.e. the peaks in the reconstructed waveform can actually be slightly higher than the highest peak values in the song file. This can cause clipping when resampling/converting especially with lossy audio formats like mp3. So to err on the safe side...

 

[John Pihel]

Another one of those threads...

Lets talk true peaks, for example why does spotify need as much as -1db true peak? Or even -2db, when mastered louder than -14Lufs. Is that really a hard rule to follow? Or do you give a d@mn about it?

I don't understand these things, I only create music

This is empty and unnecessary talk. These are their requirements and that's it. By the way, on YouTube they are the same as on Spotify. If you don't want platforms to change anything in your sound, stick to their requirements. Amen.

 

[SwitchBack]

Hmm, don't entirely agree. Understand the rules to bend the rules

 

[John Pihel]

Hmm, don't entirely agree. Understand the rules to bend the rules

Your choice. Although what is there to understand... Maybe you have at least one logical explanation, why would you do this?
I have neither the time, nor the desire, nor the reason to do this. And I can't imagine who needs it. Rules are rules. If you are not satisfied with this, pass by this site. Or do not follow their rules, but then do not be surprised that your music there will sound a little differently than you intended. Although if there is nothing else to do.....

 

[SwitchBack]

Time can be a reason to simply do what others tell you to do, I give you that. But understanding the reasons behind rules will allow you to separate the strict ones from the guidelines, and the ones holding you back from the ones keeping you from harm. In the case of peak levels it's probably the latter, but there is comfort in knowing that, right?

 

[John Pihel]

Time can be a reason to simply do what others tell you to do, I give you that. But understanding the reasons behind rules will allow you to separate the strict ones from the guidelines, and the ones holding you back from the ones keeping you from harm. In the case of peak levels it's probably the latter, but there is comfort in knowing that, right?

Okay, great, good luck. Don't forget to share your discoveries with us.

 

[TDF1981]

Time can be a reason to simply do what others tell you to do, I give you that. But understanding the reasons behind rules will allow you to separate the strict ones from the guidelines, and the ones holding you back from the ones keeping you from harm. In the case of peak levels it's probably the latter, but there is comfort in knowing that, right?

The reason might be that theoretically with LUFS you could deliver a file that is basically silent for some time and then goes to 0dBFS for some time - if the track is long enough it will still read -14 LUFS or less. So when you look at your LUFS only you will not have any idea of temporary loudness. It‘s probably set to -1dB because some algorithms are not so good at dealing with peaks too close to 0dB.
I‘d follow those rules, it’s not that your master would sound any better at all.

 

[SwitchBack]

True peak and LUFS are two completely different parameters, with completely different rules+reasons.

The peaks are a technical matter where living too close to 0dBFS can render the file uneditable because clipping (e.g. when converting to a lossy streaming format) has become inevitable.

The maximum LUFS level is a housekeeping thing, like valet parking: "If you don't do it yourself then we will do it for you." As long as the song is properly mastered with the appropriate dynamic range (limit) for the intended use then I would go in hot-ish (respecting the true peak limit). They (the station/streaming service) will determine the integrated LUFS, most likely just put the value in the file header, and use that to turn it down to their across-songs level on playback. And then the listener can turn it up again

 

[nikkdust]

The reason might be that theoretically with LUFS you could deliver a file that is basically silent for some time and then goes to 0dBFS for some time - if the track is long enough it will still read -14 LUFS or less. So when you look at your LUFS only you will not have any idea of temporary loudness. It‘s probably set to -1dB because some algorithms are not so good at dealing with peaks too close to 0dB.
I‘d follow those rules, it’s not that your master would sound any better at all.

Yes, but its -2db if you master louder that 14 lufs.

 

[SwitchBack]

I think the misconception is the 'mastering to LUFS'. Mastering is to a chosen dynamic range, highest peaks to quietest sections also with end use in mind: a home theater set can handle a wider dynamic range than what's useable in your car. Integrated LUFS is a weighted average, a level you can dial in with the master fader after all else is done, provided that the end result doesn't exceed true peak limits. When mastering (music) is done correctly then with 99 out of 100 songs the true peak value will be far below 0dBFS after -14LUFS is set.

 

[Craig Anderton]

I was talking with an award-winning mastering engineer who said True Peak is "technical BS" because he couldn't hear it when he was mastering. I tried to explain that you'll mostly hear it under two circumstances: during playback after the signal passes through a D/A converter, or when transcoding audio to a compressed format (which is the relevance with streaming services). His attitude was "I don't hear it while mastering, so you're an idiot to believe in this technical BS." Uh...okay. I also tried to explain that the sonic results were different in the two scenarios, but by that time he just assumed I knew nothing, and how dare I consider myself a mastering engineer

At one point I did a test to find out the audible effect excessive true peak values had when converted to 320 kbps mp3. It didn't sound like conventional distortion, it was more of a fuzzy quality, like a super-subtle version of a blown speaker. It's worth following "the rules" to avoid that.

 

[SwitchBack]

I was talking with an award-winning mastering engineer who said True Peak is "technical BS" because he couldn't hear it when he was mastering. I tried to explain that you'll mostly hear it under two circumstances: during playback after the signal passes through a D/A converter, or when transcoding audio to a compressed format (which is the relevance with streaming services). His attitude was "I don't hear it while mastering, so you're an idiot to believe in this technical BS." Uh...okay. I also tried to explain that the sonic results were different in the two scenarios, but by that time he just assumed I knew nothing, and how dare I consider myself a mastering engineer

At one point I did a test to find out the audible effect excessive true peak values had when converted to 320 kbps mp3. It didn't sound like conventional distortion, it was more of a fuzzy quality, like a super-subtle version of a blown speaker. It's worth following "the rules" to avoid that.

Haha, yeah, a lot of religious rationalists in sound engineering It does touch on an (imho) interesting philosophical question: Is sound engineering/mastering mostly a craft or an art? A craft suggests a trade learned from or by observing the masters, honed by practice, performed by rules and traditions set in the past. An art suggests challenging the rules and going where no man has gone before. With sound engineering as a craft I could see TP never becoming an issue because properly done nothing will 'stick out' in the first place. But with new technologies boundaries move (for better or for worse) and new boundaries appear. Sound engineering as an art will challenge those boundaries and redefine them. And in the end that's how the craft evolves too

 

[John Pihel]

Haha, yeah, a lot of religious rationalists in sound engineering It does touch on an (imho) interesting philosophical question: Is sound engineering/mastering mostly a craft or an art? A craft suggests a trade learned from or by observing the masters, honed by practice, performed by rules and traditions set in the past. An art suggests challenging the rules and going where no man has gone before. With sound engineering as a craft I could see TP never becoming an issue because properly done nothing will 'stick out' in the first place. But with new technologies boundaries move (for better or for worse) and new boundaries appear. Sound engineering as an art will challenge those boundaries and redefine them. And in the end that's how the craft evolves too

Maybe, in addition to this useless and amateurish blah blah blah, you will demonstrate your achievements in sound engineering as an "artist" and not a craftsman.
For more than 40 years of working as a sound engineer, I have heard enough of these "artists". As a rule, these are amateurs who do not know how to mix music correctly and justify their ignorance with "artistic intent".
Surprise us. Good luck.
Maybe then I will take your statements seriously.

 

[SwitchBack]

Maybe, in addition to this useless and amateurish blah blah blah, you will demonstrate your achievements in sound engineering as an "artist" and not a craftsman.
For more than 40 years of working as a sound engineer, I have heard enough of these "artists". As a rule, these are amateurs who do not know how to mix music correctly and justify their ignorance with "artistic intent".
Surprise us. Good luck.
Maybe then I will take your statements seriously.

So, in your 40 years of working as a sound engineer, did you ever change your approach to mixing/mastering?

 

[Craig Anderton]

But with new technologies boundaries move (for better or for worse) and new boundaries appear. Sound engineering as an art will challenge those boundaries and redefine them. And in the end that's how the craft evolves too

Robert Frank said "There is nothing worse than a brilliant image of a fuzzy concept." Subjective opinion alert: I'd rather hear imperfectly mixed but innovative music than music that's perfectly crafted but has no innovation. However, I'd MUCH rather hear music that incorporates both.

Art, like beauty, is in the eye of the beholder. That's why Van Gogh was a resounding commercial failure during his lifetime, and now his paintings are worth millions of dollars. His paintings didn't change - the perception of what constitutes "art" changed. You could just as easily cite Stravinsky, Coltrane, and others who challenged boundaries and were excoriated for it, only to have their genius recognized decades, or even centuries, later.

Of course, it probably goes without saying that artists on the level of Van Gogh, Stravinsky, or Coltrane are few and far between.

 

[John Pihel]

So, in your 40 years of working as a sound engineer, did you ever change your approach to mixing/mastering?

The approach to mixing/mastering changes constantly depending on the material recorded in the studio. Nothing more. You can do "artistic experiments" at home and not in the studio on someone else's material. That's why I wrote that I expect a miracle from you.

Craig, the cases you mentioned are just exceptions to the rule. As you yourself noted, this is a great rarity.

 

[Craig Anderton]

You can do "artistic experiments" at home and not in the studio on someone else's material.

But bear in mind that sometimes the musician wants artistic input, not just technical input. Most artists have been very open to seeing what I can do. Sometimes they like it, sometimes they don't, but they like the option to see their work from a different artistic standpoint.

 

[John Pihel]

But bear in mind that sometimes the musician wants artistic input, not just technical input. Most artists have been very open to seeing what I can do. Sometimes they like it, sometimes they don't, but they like the option to see their work from a different artistic standpoint.

Well, naturally. If the musicians KNOW what they want, that's good and naturally I tried to do it that way. After all, if not, they can go to another studio or change the sound engineer, etc. I haven't had such a case.
But in any case, purely technically, the result should be CORRECT, including the requirements of the platform where they are going to publish their music. This is an AXIOM.

P.S. I have been retired for 10 years and under the name John Pihel I write songs for myself and friends. My last place of work was Hardstudios in Switzerland. My first internship was Abbey Road in the late seventies. Before that, I graduated from the Royal Academy of Music in the UK.

I am writing a book about sound engineering. I finally have some free time. I didn't have any at all when I was working.

P.S.P.S. When my old friend Bob Katz (we became friends when I was working in LA) published his book, I asked him - do you have free time for this, buddy? He said - I don't sleep more than four hours. And he didn't make a damn thing on it. This is some kind of fanaticism.

 

[lokeyfly]

Not sure where this thread is going, but in the words of Jackie Mason during a Honda car commercial [thick Yiddish accent] he says: "The front wheels go like this, and the rear wheels go like this (other way), and before you know it, you're parked. And why do you need this? .............. I don't know."